On the Inference Calibration of Neural Machine Translation

Shuo Wang, Zhaopeng Tu, Shuming Shi, Yang Liu


Abstract
Confidence calibration, which aims to make model predictions equal to the true correctness measures, is important for neural machine translation (NMT) because it is able to offer useful indicators of translation errors in the generated output. While prior studies have shown that NMT models trained with label smoothing are well-calibrated on the ground-truth training data, we find that miscalibration still remains a severe challenge for NMT during inference due to the discrepancy between training and inference. By carefully designing experiments on three language pairs, our work provides in-depth analyses of the correlation between calibration and translation performance as well as linguistic properties of miscalibration and reports a number of interesting findings that might help humans better analyze, understand and improve NMT models. Based on these observations, we further propose a new graduated label smoothing method that can improve both inference calibration and translation performance.
Anthology ID:
2020.acl-main.278
Volume:
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics
Month:
July
Year:
2020
Address:
Online
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
3070–3079
Language:
URL:
https://aclanthology.org/2020.acl-main.278
DOI:
10.18653/v1/2020.acl-main.278
Bibkey:
Copy Citation:
PDF:
https://aclanthology.org/2020.acl-main.278.pdf
Video:
 http://slideslive.com/38929377